A routine trip to the emergency department may be one of the most underused opportunities in modern medicine to change a patient’s life for the better. New research from Hong Kong suggests that a short phone call followed by months of supportive text messages can nudge discharged emergency patients away from the behaviors that quietly fuel the world’s deadliest diseases. The randomized controlled trial, published in PLOS Medicine, found that nearly a third of patients who received a digitally augmented health promotion intervention reported abstaining from at least one major health-risk behavior six months later, compared with one in five who received only brief standard advice.
The stakes behind this seemingly modest shift are enormous. Noncommunicable diseases such as heart disease, stroke, cancer, diabetes, and chronic respiratory illness are now the leading cause of death worldwide, and their roots lie in a handful of modifiable behaviors: tobacco use, harmful alcohol consumption, unhealthy diets, and physical inactivity. These are not exotic risks confined to wealthy nations; they accumulate silently across populations and health systems, and clinicians have long struggled to find scalable ways to intervene. Emergency departments, which see millions of patients annually at moments when health concerns are fresh in their minds, have been proposed as natural intervention points precisely because patients may be unusually receptive to advice at discharge.
The trial, led by Ho Cheung William Li and colleagues, enrolled adults aged 18 to 65 who visited a major public hospital emergency department in Hong Kong, were triaged as semi-urgent or non-urgent, reported at least one health-risk behavior, and owned a smartphone. Of 2,134 patients screened, 572 met the criteria and were randomized in equal numbers, 286 per group, to either the intervention or a control condition. The study was assessor-blinded, meaning the researchers who measured outcomes did not know which arm participants belonged to, and it was registered with ClinicalTrials.gov under number NCT06077565.
The intervention itself was built on an explicit behavioral framework rather than ad hoc counseling. Participants in the intervention arm received a brief telephone-based consultation structured around the AWARD model, an acronym standing for Ask, Warn, Advise, Refer, and Do-it-again. In practical terms, the counselor asked about the patient’s risk behaviors, warned about their consequences, advised on specific changes, referred the patient to appropriate resources, and arranged follow-up. Crucially, the phone call was then reinforced by weekly messages delivered through WhatsApp or WeChat, the dominant instant messaging platforms in Hong Kong, continuing for six months. Control participants received only brief telephone advice at discharge, reflecting what many busy emergency departments could realistically offer.
The primary outcome was self-reported abstinence from at least one health-risk behavior at six months. Here the results favored the intervention clearly: 30.1 percent of intervention participants achieved abstinence from at least one behavior, compared with 19.9 percent of controls, a relative risk of 1.51 with a 95 percent confidence interval of 1.13 to 2.02 and a p-value of 0.006. In a field where behavioral interventions often produce marginal effects, a roughly fifty percent relative improvement is noteworthy, particularly given how light-touch the intervention was. The approach required no new clinic visits, no specialist staff on site, and no expensive technology beyond the phones patients already carried.
Secondary outcomes reinforced the picture. The intervention significantly increased the likelihood that participants had reduced the number of risky behaviors they engaged in, both at six months, with a relative risk of 1.54 and a p-value of 0.01, and at twelve months, with a relative risk of 1.48 and a p-value of 0.02. Physical inactivity showed the single largest improvement at the six-month mark: 31.7 percent of intervention participants reported abstinence from sedentary behavior compared with 16.2 percent of controls, a difference the authors reported as statistically significant at p less than 0.001. That finding is intriguing because physical activity may be the easiest of the four target behaviors to modify without clinical supervision, requiring no prescription, substitution therapy, or structured program, only a change in daily routine that supportive messaging can repeatedly encourage.
Yet the trial’s most sobering result emerged at the twelve-month assessment. The advantage the intervention had enjoyed at six months attenuated after the weekly booster messaging stopped, and the difference in abstinence between the groups was no longer statistically significant for the primary outcome. In other words, the behavior change, while real, appears to have depended on the continued presence of the digital support. Once the messages ceased, many participants appear to have drifted back toward old habits, a pattern all too familiar to anyone who has studied smoking cessation, weight loss, or exercise adherence. The finding does not invalidate the approach, but it reframes it: the intervention works while it is working, and sustaining gains likely requires either extended messaging, periodic booster cycles, or strategies that build self-sustaining motivation before support is withdrawn.
The trial’s design choices and limitations deserve careful attention when weighing these results. All outcomes relied on self-report, which is vulnerable to social desirability bias, particularly in a study where participants knew they were being asked about health behaviors. The single-center design in one Hong Kong hospital raises questions about generalizability to other health systems, cultures, and messaging platforms, although the use of WhatsApp and WeChat covers a substantial share of the world’s smartphone users. Loss to follow-up, an inevitable feature of trials that track discharged patients over a year, may also have influenced the estimates. The authors themselves call for multicenter trials with longer follow-up to establish whether the benefits can be preserved and reproduced at scale.
Even with those caveats, the study carries a message that health systems worldwide should find hard to ignore. Emergency departments are often criticized as expensive, overcrowded endpoints of failed prevention, treating the consequences of behaviors that cheaper interventions could have addressed years earlier. This trial suggests the ED visit itself can be converted into a teachable moment at remarkably low cost. A brief structured phone call, a behavioral model any trained counselor can follow, and six months of automated weekly messages produced measurable change in a population that standard discharge advice largely leaves untouched. If the messaging infrastructure is already in place, the marginal cost of extending support to additional patients is small, which is exactly the kind of arithmetic that population-level prevention demands.
The broader implication is that digital augmentation may be the bridge between the fleeting contact of acute care and the sustained effort that behavior change requires. Emergency clinicians rarely have time to counsel patients on diet, exercise, smoking, and drinking, and patients rarely retain much of what is said in the chaos of discharge. By shifting the sustained portion of the intervention onto the devices in patients’ pockets, the Hong Kong team effectively decoupled the intervention from the hospital walls. The twelve-month attenuation shows the bridge is not yet long enough, but the six-month results show it can carry real weight. Future iterations, with longer messaging schedules, adaptive content that responds to relapse, or integration with primary care follow-up, will determine whether a text message thread begun in a crowded emergency room can bend the trajectory of the chronic diseases that now dominate global mortality.
Subject of Research: A digitally augmented health promotion intervention for reducing health-risk behaviors in emergency department discharge patients
Article Title: Effect of a digitally augmented general health promotion intervention on abstinence from health-risk behaviors among emergency department discharge patients: A randomized controlled trial
Article References: Li, H. C. W., Xia, W., Yeung, J., Ho, L. L. K., Cheung, T. A., Cheung, K. Y., Chan, Y. C., Chan, W. Y., Lui, C. W., & Chen, H. (2026). Effect of a digitally augmented general health promotion intervention on abstinence from health-risk behaviors among emergency department discharge patients: A randomized controlled trial. PLOS Medicine, 23(9), e1004916. https://doi.org/10.1371/journal.pmed.1004916
Image Credits: AI Generated
DOI: 10.1371/journal.pmed.1004916
Keywords: emergency department, health-risk behaviors, randomized controlled trial, digital health intervention, noncommunicable diseases, physical inactivity, WhatsApp messaging, health promotion, behavior change, PLOS Medicine, Hong Kong, AWARD model
News Source: Ophelia Keating. (October 10, 2026). Text Messages After the ER Help Patients Quit Risky Habits, Trial Finds. Scienmag.
